Switched reluctance motor power converter fault detection and diagnosis is mainly a qualitative analysis method based on a system model. By modeling and simulation or experiment method, it analyzes and compares the change information of observed quantities of the system in normal and fault states, and extracts residual error or fault features with a view to realize fault isolation of the system by fault decision algorithm. For a system a mathematical model of which can be accurately established, this method can be used to intensively study system fault mechanism. It not only makes for feature analysis of known faults, but also may conduct performance analysis and diagnosis for the faults which have never appeared or for which the experience is inadequate, so it is widely applied. Switched reluctance motor has a double salient structure and non-sine power supply features different from a conventional AC-DC motor. Consequently, the fault isolation and fault feature extraction of switched reluctance motor power converter are different from other motors obviously. When qualitative analysis method based on system model is applied in fault diagnosis of switched reluctance motor power converters, the difficulty rests with the establishment of an accurate mathematical model for switched reluctance motor system. The fault diagnosis method with no need for an accurate mathematical model for switched reluctance motor system is one of the important directions of the current research and technical development of switched reluctance motor system.